Enhanced Breast Cancer Prediction using ML: Comparative Analysis and Optimized Models

Medical research continues to focus on breast cancer diagnosis and its prognostic methods since it remains a leading death cause worldwide. Breast cancer prediction receives an enhancement through this study which implements a comparative analysis of machine learning models. Five classifiers, namely KNN, SVC, RF, DT, and Gradient Boosting, were applied to analyse the Wisconsin Breast Cancer Diagnostic (WBCD) dataset. A pre-process stage includes both scaling and feature selection to enhance the optimisation of model performance. Models are evaluated through assessments of accuracy as well as precision and recall, and F1-score. The Gradient Boosting Classifier functions as the optimal predictor because it displays an accuracy rate of 99% above all other models tested. The performance of SVC and RF outmatches the precision levels of the KNN and DT classifiers. The classification process requires the selection of an optimal ML model for breast cancer detection because researchers need to combine multiple models and better understand the effects of feature engineering. The research supports active ML diagnostic tool development which helps healthcare professionals detect breast cancer earlier while making improved clinical choices for patient treatment.